Simultaneous reconstruction of emission activity and attenuation coefficient distribution from TOF data, acquired with external shell source

1. A method for generating images from time of flight (TOF) data associated with a scan of at least one object using a positron emission tomography system, comprising: providing initial values for an activity image to yield a current activity image; estimating initial values for an attenuation map (μ-map) image based on the TOF data to yield a current μ-map image; repeating, until at least one termination condition is met, the steps of: updating the current activity image based on at least the current μ-map and a first update algorithm, and updating the current μ-map image based on at least on the updated activity image and a second update algorithm, outputting an image of the at least one object based on the current μ-map and the current activity image.

2. The method of claim 1 , wherein the estimating of the initial values for the μ-map comprises performing a TX data separation of the TOF data to yield non-TOF data and a TX scatter estimation.

3. The method of claim 1 , wherein the providing of the initial values for the activity image comprises selecting a uniform activity image.

4. The method of claim 1 , wherein the updating of the current μ-map image comprises selecting an update algorithm representing a scaled gradient ascent algorithm.

5. The method of claim 1 , wherein the updating of the current μ-map image comprises selecting as the second update algorithm: ⁢ μ j ( n , 0 ) = μ j ( n - 1 ) ⁢ A i ( n - 1 ) = ⅇ - ∑ j ⁢ C ij ⁢ μ j ( n - 1 ) , ⁢ ⁢ B i , t TX = B i , t + N i - 1 ⁢ ∑ j ⁢ C ij , t ⁢ f j ( n ) , b _ i , t TX = r _ i + N i - 1 ⁢ s _ i , t ⁡ ( f ) + s _ i , t ⁡ ( B ) . ⁢ μ j ( n , k ) = μ j ( n , k - 1 ) + α ⁢ ∑ i ⁢ C ij ⁢ { ∑ t ⁢ B i , t TX ⁢ A i ( n - 1 ) ( 1 - p i , t B i , t TX ⁢ A i ( n - 1 ) + b _ i , t TX ) } - β ⁢ ∂ U ⁡ ( μ ( n , k - 1 ) ) ∂ μ j D ⁢ ∑ i ⁢ C ij ⁢ { ∑ t ⁢ ( B i , t TX ⁢ A i ( n - 1 ) ) 2 B i , t TX ⁢ A i ( n - 1 ) + b _ i , t TX } + β ⁢ ∂ 2 ⁢ U ⁡ ( μ ( n , k - 1 ) ) ∂ μ j 2 , ⁢ ⁢ k = 1 , … ⁢ , K TX ⁢ μ j ( n ) = μ j ( n , K TX ) where μ is the attenuation coefficient distribution and f is the object emission activity image, indentified at image voxel with index j, B is blank projection data, C ij and C ij,t are system matrices of non-TOF and TOF line-integral projectors, r is mean random counts, N is a normalization coefficient array, s is scatter contribution, and α is a relaxation parameter.

6. The method of claim 1 , wherein updating the current activity image selecting as the first update equation: f j ( n , 0 ) = f j ( n - 1 ) ⁢ ⁢ A i ( n - 1 ) = ⅇ - ∑ j ⁢ C ij ⁢ μ j ( n - 1 ) , ⁢ b _ i , t EM = r _ i + N i - 1 ⁢ s _ i , t ⁡ ( f ) + s _ i , t ⁡ ( B ) + B i , t ⁢ A i ( n - 1 ) . f j ( n , k ) = f j ( n , k - 1 ) ⁢ ∑ i , t ⁢ C ij , t ⁢ p i , t ∑ j ′ ⁢ C ij ′ , t ⁢ f j ′ ( n , k - 1 ) + A i ( n - 1 ) - 1 ⁢ N i ⁢ b _ i , t EM ∑ i , t ⁢ C ij , t ⁢ A i ( n - 1 ) ⁢ N i - 1 , ⁢ k = 1 , … ⁢ , K EM ⁢ ⁢ f j ( n ) = f j ( n - 1 , K EM ) where μ is the attenuation coefficient distribution and f is the object emission activity image, indentified at image voxel with index j, B is blank projection data, C ij and C ij,t are system matrices of non-TOF and TOF line-integral projectors, r is mean random counts, N is a normalization coefficient array, and s is scatter contribution.

7. The method of claim 1 , the step of repeating further comprising: evaluating, for at least one of the current μ-map image and the current activity image, a difference between a current version and an updated version; if the difference for at least one of the current μ-map image and the current activity image meets a predefined termination criteria, determining that the termination condition has been met.

8. A system for generating images from time of flight (TOF) data associated with a scan of at least one object using a positron emission tomography system, comprising: a processor; and a computer-readable medium having stored thereon instructions for causing the processor to perform the steps of: providing initial values for an activity image to yield a current activity image; estimating initial values for an attenuation map (μ-map) image based on the TOF data to yield a current μ-map image; repeating, until at least one termination condition is met, the steps of: updating the current activity image based on at least the current μ-map and a first update algorithm, and updating the current μ-map image based on at least on the updated activity image and a second update algorithm, outputting an image of the at least one object based on the current μ-map and the current activity image.

9. The system of claim 8 , wherein the instructions for the estimating the initial values for the μ-map further comprise instructions for performing a TX data separation of the TOF data to yield non-TOF data and a TX scatter estimation.

10. The system of claim 8 , wherein the instructions for the providing of the initial values for the activity image further comprise instructions for selecting a uniform activity image.

11. The system of claim 8 , wherein the instructions for the updating of the current μ-map image further comprise instructions for selecting an update algorithm representing a scaled gradient ascent algorithm.

12. The system of claim 8 , wherein the instructions for the updating of the current μ-map image further comprise instructions for implementing the second update algorithm using: ⁢ μ j ( n , 0 ) = μ j ( n - 1 ) ⁢ A i ( n - 1 ) = ⅇ - ∑ j ⁢ C ij ⁢ μ j ( n - 1 ) , ⁢ ⁢ B i , t TX = B i , t + N i - 1 ⁢ ∑ j ⁢ C ij , t ⁢ f j ( n ) , b _ i , t TX = r _ i + N i - 1 ⁢ s _ i , t ⁡ ( f ) + s _ i , t ⁡ ( B ) . ⁢ μ j ( n , k ) = μ j ( n , k - 1 ) + α ⁢ ∑ i ⁢ C ij ⁢ { ∑ t ⁢ B i , t TX ⁢ A i ( n - 1 ) ( 1 - p i , t B i , t TX ⁢ A i ( n - 1 ) + b _ i , t TX ) } - β ⁢ ∂ U ⁡ ( μ ( n , k - 1 ) ) ∂ μ j D ⁢ ∑ i ⁢ C ij ⁢ { ∑ t ⁢ ( B i , t TX ⁢ A i ( n - 1 ) ) 2 B i , t TX ⁢ A i ( n - 1 ) + b _ i , t TX } + β ⁢ ∂ 2 ⁢ U ⁡ ( μ ( n , k - 1 ) ) ∂ μ j 2 , ⁢ ⁢ k = 1 , … ⁢ , K TX ⁢ μ j ( n ) = μ j ( n , K TX ) where μ is the attenuation coefficient distribution and f is the object emission activity image, identified at image voxel with index j, B is blank projection data, C ij and C ij,t are system matrices of non-TOF and TOF line-integral projectors, r is mean random counts, N is a normalization coefficient array, s is scatter contribution, and α is a relaxation parameter.

13. The system of claim 8 , wherein the instructions for the updating of the current activity image further comprise instructions for implementing the first update algorithm using: f j ( n , 0 ) = f j ( n - 1 ) ⁢ ⁢ A i ( n - 1 ) = ⅇ - ∑ j ⁢ C ij ⁢ μ j ( n - 1 ) , ⁢ b _ i , t EM = r _ i + N i - 1 ⁢ s _ i , t ⁡ ( f ) + s _ i , t ⁡ ( B ) + B i , t ⁢ A i ( n - 1 ) . f j ( n , k ) = f j ( n , k - 1 ) ⁢ ∑ i , t ⁢ C ij , t ⁢ p i , t ∑ j ′ ⁢ C ij ′ , t ⁢ f j ′ ( n , k - 1 ) + A i ( n - 1 ) - 1 ⁢ N i ⁢ b _ i , t EM ∑ i , t ⁢ C ij , t ⁢ A i ( n - 1 ) ⁢ N i - 1 , ⁢ k = 1 , … ⁢ , K EM ⁢ ⁢ f j ( n ) = f j ( n - 1 , K EM ) where μ is the attenuation coefficient distribution and f is the object emission activity image, indentified at image voxel with index j, B is blank projection data, C ij and C ij,t are system matrices of non-TOF and TOF line-integral projectors, r is mean random counts, N is a normalization coefficient array, and s is scatter contribution.

14. The system of claim 8 , the steps further comprising: evaluating, for at least one of the current μ-map image and the current activity image, a difference between a current version and an updated version; if the difference for at least one of the current μ-map image and the current activity image meets a predefined termination criteria, determining that the termination condition has been met.

15. A non-transitory computer-readable medium having stored thereon a computer program with a plurality of instructions for causing a computing device to perform the method comprising: providing initial values for an activity image to yield a current activity image; estimating initial values for an attenuation map (μ-map) image based on the TOF data to yield a current μ-map image; repeating, until at least one termination condition is met, the steps of: updating the current activity image based on at least the current μ-map and a first update algorithm, and updating the current μ-map image based on at least on the updated activity image and a second update algorithm, outputting an image of the at least one object based on the current μ-map and the current activity image.

16. The non-transitory computer-readable medium of claim 15 , wherein determining the initial values for the μ-map comprises performing a TX data separation of the TOF data to yield non-TOF data and a TX scatter estimation.

17. The non-transitory computer-readable medium of claim 15 , wherein the updating of the current μ-map image comprises selecting an update algorithm representing a scaled gradient ascent algorithm.

18. The non-transitory computer-readable medium of claim 15 , wherein the updating of the current μ-map image comprises selecting as the second update algorithm: ⁢ μ j ( n , 0 ) = μ j ( n - 1 ) ⁢ A i ( n - 1 ) = ⅇ - ∑ j ⁢ C ij ⁢ μ j ( n - 1 ) , ⁢ ⁢ B i , t TX = B i , t + N i - 1 ⁢ ∑ j ⁢ C ij , t ⁢ f j ( n ) , b _ i , t TX = r _ i + N i - 1 ⁢ s _ i , t ⁡ ( f ) + s _ i , t ⁡ ( B ) . ⁢ μ j ( n , k ) = μ j ( n , k - 1 ) + α ⁢ ∑ i ⁢ C ij ⁢ { ∑ t ⁢ B i , t TX ⁢ A i ( n - 1 ) ( 1 - p i , t B i , t TX ⁢ A i ( n - 1 ) + b _ i , t TX ) } - β ⁢ ∂ U ⁡ ( μ ( n , k - 1 ) ) ∂ μ j D ⁢ ∑ i ⁢ C ij ⁢ { ∑ t ⁢ ( B i , t TX ⁢ A i ( n - 1 ) ) 2 B i , t TX ⁢ A i ( n - 1 ) + b _ i , t TX } + β ⁢ ∂ 2 ⁢ U ⁡ ( μ ( n , k - 1 ) ) ∂ μ j 2 , ⁢ ⁢ k = 1 , … ⁢ , K TX ⁢ μ j ( n ) = μ j ( n , K TX ) where μ is the attenuation coefficient distribution and f is the object emission activity image, indentified at image voxel with index j, B is blank projection data, C ij and C ij,t are system matrices of non-TOF and TOF line-integral projectors, r is mean random counts, N is a normalization coefficient array, s is scatter contribution, and α is a relaxation parameter.

19. The non-transitory computer-readable medium of claim 15 , wherein updating the current activity image selecting as the first update equation: f j ( n , 0 ) = f j ( n - 1 ) ⁢ ⁢ A i ( n - 1 ) = ⅇ - ∑ j ⁢ C ij ⁢ μ j ( n - 1 ) , ⁢ b _ i , t EM = r _ i + N i - 1 ⁢ s _ i , t ⁡ ( f ) + s _ i , t ⁡ ( B ) + B i , t ⁢ A i ( n - 1 ) . f j ( n , k ) = f j ( n , k - 1 ) ⁢ ∑ i , t ⁢ C ij , t ⁢ p i , t ∑ j ′ ⁢ C ij ′ , t ⁢ f j ′ ( n , k - 1 ) + A i ( n - 1 ) - 1 ⁢ N i ⁢ b _ i , t EM ∑ i , t ⁢ C ij , t ⁢ A i ( n - 1 ) ⁢ N i - 1 , ⁢ k = 1 , … ⁢ , K EM ⁢ ⁢ f j ( n ) = f j ( n - 1 , K EM ) where μ is the attenuation coefficient distribution and f is the object emission activity image, indentified at image voxel with index j, B is blank projection data, C ij and C ij,t are system matrices of non-TOF and TOF line-integral projectors, r is mean random counts, N is a normalization coefficient array, and s is scatter contribution.

20. The non-transitory computer-readable medium of claim 15 , the step of repeating further comprising: evaluating, for at least one of the current μ-map image and the current activity image, a difference between a current version and an updated version; if the difference for at least one of the current μ-map image and the current activity image meets a predefined termination criteria, determining that the termination condition has been met.